Newbie workflow problem multitrack vs waveform and matching loudness

[straightlife]

How do you do it???

I'm working on a podcast,  an audiocast of a book.

I read for ten minutes. Put it down. read for another ten minutes, etc.

If I do noise reduction on the various tracks and then fix the esses and the breathing IN THE WAVEFORM editor (making sure not to erase the breathing entirely) and THEN bring them all into the multitrack and try to do the loudness adjustment in the MULTI TRACK, so that all the tracks match each other, the breathing on some of the tracks and the mouth noises which were successfully reduced in the WAVEFORM, are suddenly back again!

And so I have to go back to the waveform editor and reduce them again. 

Should I START in multitrack, match the loudness, and then go to waveform to fix the breathing and mouth noises, etc. being VERY CAREFUL not to change the volume and mess up the adjustment I did in multi. Or what. Please?????

[SteveG_AudioMasters_ · Accepted solution]

Absolutely the best way to do this is to record everything in Multitrack. For a start, it's direct to disk so if anything goes wrong, you don't lose everything. The other thing you should do is not to alter anything about the way you are recording, between takes. Get it set up correctly first, so that the number of things you need to do is minimised. Then, without altering anything or doing any processing, edit the takes together in multitrack so that you have a whole segment (say, a chapter) that all sounds the same. You make a note of all the settings on everything, and then process that chapter as a whole by mixing it down to a chapter file and processing that in Waveform view. Why? Because you need to make a note of all the settings you've used to process the track (save everything as presets, or even set up a Favorite if you want). The reason is that when you come to do the next chapter, with care you can use the same presets. Yes, you have to watch out a bit with NR, but even that can be replicated if you save a noise print, and create a preset for the processing. The other important thing to note about doing it this way is that your original multitrack (unprocessed) edit remains intact, so if you need to, you have something to go back to.

Doing it this way means that you should end up with consistent results, and if you need to go back to anything, it shouldn't be too difficult. It's all about having a processing structure - once you've got that sorted out, and you can stick to it, the rest is pretty easy.

It's good that you've realised that breathing makes speech sound natural - it really does make a difference in long-form recording to have a natural-sounding voice. It means that without effort, the listener should be able to have a more 'immersive' experience - which is of course what I would have thought most people wanted from an audiobook!

[straightlife]

Oh THANK YOU for your wisdom. I'm going to try that with the next section. I always learn things by rushing in and making mistakes and then, oh! Well, I guess I won't do that anynore.

One question about what you said. I'm reading long chapters. I have to stop a few times or my voice goes. The next time I start where I left off. I should do that in the SAME multitrack?  My setup is set, but maybe it'd be good to measure my distance from the mic and use that measurement every time. 

[SteveG_AudioMasters_]

Yes it's fine to record on the same track. Audition will keep all of the files, whether you want it to or not. A lot of people don't like that, but equally, Adobe don't want to get sued when your lawyer thinks it's their fault that you weren't given the opportunity to save absolutely everything... Yes, getting in the same (comfortable) position is a good idea. So is only drinking clear liquids between takes, but not alcoholic ones - which only really leaves water, I'm afraid. Anything milky leaves membranes forming around the back of your mouth, and even fruit juices have their downsides. Meals are generally a disaster, as people generally sound different after they've eaten; we always recommend that if we're going to do a reading, it's in either a morning or afternoon session, with only comfort breaks. It's a good discipline, as generally you end up with less editing to do this way. Smoking is also a no-no unless you're specifically going for a really dodgy-sounding voice - generally not recommended, as smokers often don't have the stamina for a long spoken session.